author：JevonWei
版权声明：原创作品

Keepalive实战之LVS-DR

实验目的：构建LVS-DR架构，为了达到LVS的高可用目的，故在LVS-DR的Director端做Keepalive集群，在Director-A上做keepalive-A，在Director上做keepalive-B，LVS-RS1和LVS-RS2为后端的两台web服务器，通过在Director上做keepalive集群实现高可用的目的

网络拓扑图

实验环境(keepalive节点同时作为LVS的directory节点)

keepalive-A(Director-A) 172.16.253.108
keepalive-B(Director-A) 172.16.253.105
LVS-RS1                 172.16.250.127
LVS-RS2                 172.16.253.193
VIP                     172.16.253.150
client                  172.16.253.177

LVS-RS web集群

为了更好的观察实验结果，故在此将RS1和RS2的web页面内容设置不一致，以致可以更清晰的区分RS1服务端和RS2服务端

LVS-RS1

[root@LVS-RS1 ~]# systemctl restart chronyd  \\多台服务器时间同步
[root@LVS-RS1 ~]# iptables -F
[root@LVS-RS1 ~]# setenforce 0
[root@LVS-RS1 ~]# yum -y install nginx
[root@LVS-RS1 ~]# vim /usr/share/nginx/html/index.html
<h1> Web RS1 </h1>
[root@LVS-RS1 ~]# systemctl start nginx修改内核参数并添加VIP地址
[root@LVS-RS1 ~]# vim lvs_dr.sh
#!/bin/bash
#
vip=172.16.253.150
mask=255.255.255.255
iface="lo:0"case $1 in
start)ifconfig $iface $vip netmask $mask broadcast $vip uproute add -host $vip dev $ifaceecho 1 > /proc/sys/net/ipv4/conf/all/arp_ignoreecho 1 > /proc/sys/net/ipv4/conf/lo/arp_ignoreecho 2 > /proc/sys/net/ipv4/conf/all/arp_announceecho 2 > /proc/sys/net/ipv4/conf/lo/arp_announce;;
stop)echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignoreecho 0 > /proc/sys/net/ipv4/conf/lo/arp_ignoreecho 0 > /proc/sys/net/ipv4/conf/all/arp_announceecho 0 > /proc/sys/net/ipv4/conf/lo/arp_announceifconfig $iface down;;
*)echo "Usage:$(basename $0) start|stop"exit 1;;
esac
[root@LVS-RS1 ~]# bash lvs_dr.sh start
[root@LVS-RS1 ~]# ifconfig
lo:0: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536inet 172.16.253.150  netmask 255.255.255.255loop  txqueuelen 1  (Local Loopback)

LVS-RS2

[root@LVS-RS2 ~]# systemctl restart chronyd  \\多台服务器时间同步
[root@LVS-RS2 ~]# iptables -F
[root@LVS-RS2 ~]# setenforce 0
[root@LVS-RS2 ~]# yum -y install nginx
[root@LVS-RS2 ~]# vim /usr/share/nginx/html/index.html
<h1> Web RS2 </h1>
[root@LVS-RS2 ~]# systemctl start nginx修改内核参数并添加VIP地址
[root@LVS-RS2 ~]# vim lvs_dr.sh
#!/bin/bash
#
vip=172.16.253.150
mask=255.255.255.255
iface="lo:0"case $1 in
start)ifconfig $iface $vip netmask $mask broadcast $vip uproute add -host $vip dev $ifaceecho 1 > /proc/sys/net/ipv4/conf/all/arp_ignoreecho 1 > /proc/sys/net/ipv4/conf/lo/arp_ignoreecho 2 > /proc/sys/net/ipv4/conf/all/arp_announceecho 2 > /proc/sys/net/ipv4/conf/lo/arp_announce;;
stop)echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignoreecho 0 > /proc/sys/net/ipv4/conf/lo/arp_ignoreecho 0 > /proc/sys/net/ipv4/conf/all/arp_announceecho 0 > /proc/sys/net/ipv4/conf/lo/arp_announceifconfig $iface down;;
*)echo "Usage:$(basename $0) start|stop"exit 1;;
esac
[root@LVS-RS1 ~]# bash lvs_dr.sh start
[root@LVS-RS1 ~]# ifconfig
lo:0: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536inet 172.16.253.150  netmask 255.255.255.255loop  txqueuelen 1  (Local Loopback)

Keepalive集群

Director节点搭建

keepalive-A

[root@keepaliveA ~]# systemctl restart chronyd  \\多台服务器时间同步
[root@keepaliveA ~]# yum -y install ipvsadm

keepalive-B

[root@keepaliveB ~]# systemctl restart chronyd  \\多台服务器时间同步
[root@keepaliveB ~]# yum -y install ipvsadm

keepalive上配置web的sorry server

keepalive-A

[root@keepaliveA ~]# yum -y install nginx
[root@keepaliveA ~]# vim /usr/share/nginx/html/index.html
</h1> sorry from Director-A(keepalive-A) </h1>
[root@keepaliveA ~]# systemctl start nginx

keepalive-B

[root@keepalive-B ~]# yum -y install nginx
[root@keepalive-B ~]# vim /usr/share/nginx/html/index.html
</h1> sorry from Director-B(keepalive-B) </h1>
[root@keepaliveB ~]# systemctl start nginx

keepalive-A配置keepalive

keepalive-A

[root@keepalive-A ~]# iptables -F
[root@keepalive-A ~]# yum -y install keepalived
[root@keepaliveA ~]# vim /etc/keepalived/keepalived.conf
global_defs {  notification_email {  \\定义邮件通知设置jevon@danran.com  \\定义邮件接收地址}notification_email_from ka_admin@danran.com \\邮件发送者smtp_server 127.0.0.1 \\邮件server服务器smtp_connect_timeout 30  \\连接超时router_id keepaliveA \\route的ID信息，自定义vrrp_mcast_group4 224.103.5.5 \\多播地址段，默认为224.0.0.18
}
vrrp_instance VI_A {state MASTERinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass qr8hQHuL}
virtual_ipaddress {172.16.253.150/32 dev ens33
}virtual_server 172.16.253.150 80 {delay_loop 6  \\服务轮询的时间间隔lb_algo rr  \\定义调度方法；lb_kind DR  \\集群的类型；protocol TCP \\服务协议，仅支持TCP；sorry_server 127.0.0.1 80 \\指定sorry server，且为本机的wen服务提供的web页面real_server 172.16.250.127 80 {weight 1  \\权重SSL_GET {  \\应用层检测url {path /  \\定义要监控的URL#digest ff20ad2481f97b1754ef3e12ecd3a9cc \\判断上述检测机制为健康状态的响应的内容的校验码；status_code 200  \\判断上述检测机制为健康状态的响应码}connect_timeout 3 \\连接请求的超时时长；nb_get_retry 3  \\重试次数delay_before_retry 1  \\重试之前的延迟时长}}real_server  172.16.253.193 80 {weight 1SSL_GET {url {path /#digest ff20ad2481f97b1754ef3e12ecd3a9cc status_code 200}connect_timeout 3nb_get_retry 3delay_before_retry 1}}
}
[root@keepaliveA ~]# systemctl start keepalived
[root@keepaliveA ~]# ip a l
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:75:dc:3c brd ff:ff:ff:ff:ff:ffinet 172.16.253.150/32 scope global ens33valid_lft forever preferred_lft forever
[root@keepaliveA ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  172.16.253.150:80 rr-> 172.16.250.127:80            Route   1      0          0
-> 172.16.253.193:80            Route   1      0          0   

keepalive-B配置keepalive

keepalive-B

[root@keepalive-B ~]# iptables -F
[root@keepalive-B ~]# yum -y install keepalived
[root@keepaliveA ~]# vim /etc/keepalived/keepalived.conf
global_defs {  notification_email {  \\定义邮件通知设置jevon@danran.com  \\定义邮件接收地址}notification_email_from ka_admin@danran.com \\邮件发送者smtp_server 127.0.0.1 \\邮件server服务器smtp_connect_timeout 30  \\连接超时router_id keepaliveA \\route的ID信息，自定义vrrp_mcast_group4 224.103.5.5 \\多播地址段，默认为224.0.0.18
}
vrrp_instance VI_A {state BACKUPinterface ens33virtual_router_id 51priority 95advert_int 1authentication {auth_type PASSauth_pass qr8hQHuL}
virtual_ipaddress {172.16.253.150/32 dev ens33
}virtual_server 172.16.253.150 80 {delay_loop 6  \\服务轮询的时间间隔lb_algo rr  \\定义调度方法；lb_kind DR  \\集群的类型；protocol TCP \\服务协议，仅支持TCP；sorry_server 127.0.0.1 80 \\指定sorry server，且为本机的wen服务提供的web页面real_server 172.16.250.127 80 {weight 1  \\权重SSL_GET {  \\应用层检测url {path /  \\定义要监控的URL#digest ff20ad2481f97b1754ef3e12ecd3a9cc \\判断上述检测机制为健康状态的响应的内容的校验码；status_code 200  \\判断上述检测机制为健康状态的响应码}connect_timeout 3 \\连接请求的超时时长；nb_get_retry 3  \\重试次数delay_before_retry 1  \\重试之前的延迟时长}}real_server  172.16.253.193 80 {weight 1SSL_GET {url {path /#digest ff20ad2481f97b1754ef3e12ecd3a9cc status_code 200}connect_timeout 3nb_get_retry 3delay_before_retry 1}}
}
[root@keepaliveB ~]# systemctl start keepalived
[root@keepalive-B ~]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  172.16.253.150:http rr-> 172.16.250.127:http          Route   1      0          0         -> 172.16.253.193:http          Route   1      0          0     

访问测试

client测试

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>

当keepalive-A故障时

[root@keepaliveA ~]# systemctl stop keepalived  

keepalive-B自动成为MASTER主节点，则LVS的director调度服务器切换为keepalive-B上，LVS-RS1和LVS-RS2的web服务正常使用

client访问测试

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>

当keepalive-A修恢复正常时，keepalive-A再次成为MASTER主节点

[root@keepaliveA ~]# systemctl start keepalived
[root@keepaliveA ~]# ip a l
: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:75:dc:3c brd ff:ff:ff:ff:ff:ffinet 172.16.253.150/32 scope global ens33valid_lft forever preferred_lft forever

当LVS-RS1的web服务故障时

[root@LVS-RS1 ~]# iptables -A INPUT -p tcp --dport 80 -j REJECT 

client访问

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
<h1> Web RS2 </h1>
<h1> Web RS2 </h1>
<h1> Web RS2 </h1>
<h1> Web RS2 </h1>

当LVS-RS1和LVS-RS2的web服务全部故障时

[root@LVS-RS1 ~]# iptables -A INPUT -p tcp --dport 80 -j REJECT
[root@LVS-RS2 ~]# iptables -A INPUT -p tcp --dport 80 -j REJECT

client访问到的时sorry server服务器，且sorry server服务器为keepalive-A

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
</h1> sorry from Director-A(keepalive-A) </h1>
</h1> sorry from Director-A(keepalive-A) </h1>
</h1> sorry from Director-A(keepalive-A) </h1>
</h1> sorry from Director-A(keepalive-A) </h1>
</h1> sorry from Director-A(keepalive-A) </h1> 

当keepalive-A故障时

[root@keepaliveA ~]# systemctl stop keepalived.service

client访问sorry server服务页面，且sorry server服务器为keepalive-B

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
</h1> sorry from Director-B(keepalive-B) </h1>
</h1> sorry from Director-B(keepalive-B) </h1>
</h1> sorry from Director-B(keepalive-B) </h1>
</h1> sorry from Director-B(keepalive-B) </h1>
</h1> sorry from Director-B(keepalive-B) </h1>

LVS-RS1的web服务恢复正常后

[root@LVS-RS1 ~]# iptables -F

client访问测试

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
<h1> Web RS1 </h1>
<h1> Web RS1 </h1>
<h1> Web RS1 </h1>
<h1> Web RS1 </h1>
<h1> Web RS1 </h1>

LVS-RS1和LVS-RS2的web服务全部恢复正常后

[root@LVS-RS1 ~]# iptables -F  [root@LVS-RS2 ~]# iptables -F  

client访问测试

[root@client ~]# for i in {1..10};do curl http://172.16.253.150;done
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>
<h1> Web RS1 </h1>
<h1> Web RS2 </h1>    

保存及重载规则

保存：建议保存至/etc/sysconfig/ipvsadm

ipvsadm-save > /PATH/TO/IPVSADM_FILE
ipvsadm -S > /PATH/TO/IPVSADM_FILE
systemctl stop ipvsadm.service

重载：

ipvsadm-restore < /PATH/FROM/IPVSADM_FILE
ipvsadm -R < /PATH/FROM/IPVSADM_FILE
systemctl restart ipvsadm.service

keepalive节点通过DNS域名解析指向实现

获取web主页面内容的校验码

[root@keepaliveA ~]# genhash -s 172.16.250.127 -p 80 -u /